(Code Review) is a code review software commonly used means of development, compared with the QA test, it is easier to find and architecture as well as the timing correlation is more difficult to find the problem, can also help the team members to improve the programming skills, unified programming style etc..
1. code review requires teams to have good culture,
team needs to recognize that code reviews are designed to improve the overall team’s capabilities, rather than checking checkpoints for individual settings".
A code has a bug found by B, so A is not capable, B ability is better, this kind of trap is easy to be diffused, thus affecting collaboration within the team, so need to avoid.
in addition, code reviews themselves can enhance the developer’s ability to learn from his own mistakes and learn from other people’s ideas. If the developer has a conflict or antipathy towards the process, this goal will not be achieved.
2. prudent use of the findings in the review rate as evaluation criteria
in the code review, if found problems, for the discoverer of the problem, this is a good thing, should be encouraged. But for those found, we do not advocate the use of this method to punish. Bug is unavoidable in software development, and excessive demanding itself is contrary to common sense. To make matters worse, the code review would be of no value or significance if participants were afraid of taking responsibility and were reluctant to point out problems in the review.
3. controls the number of codes for each review
, according to smartbear’s survey at CISCO, "the 200 lines of -400 code are the best every time.". Every time you try to review too much code, the ability to spot problems falls, and the exact scaling relationship is shown in the figure below:
we have found in practice that the optimal amount of code review varies with the development platform and the development language. But limiting the number of reviews is really necessary because the process is a highly intensive, brain intensive activity. For a long time, the code in the eyes of examiners is only letters, without any logical connection, naturally there will not be too much output.
4. goes to
, in every code review, we ask reviewers to use their experience to think about problems that might come up first, and then verify that these issues have been resolved by reviewing them. One trick is to start with a user – aware feature, assuming a more complex usage scenario, and verify that the usage scenario works correctly in code reading.
uses this technique to allow censors to substitute >